Overview

All data have to be stored somewhere

Whereas we have concentrated more on algorithms, methods, and activities in the previous parts of our OOP course, this chapter discusses databases. In most applications, such as billing software, working with data is at the center of activities. These data not only are input but also have to be stored and retrieved again and again (even many months later). Think only of the postings entered in a financial accounting program.

Similarly, almost every COBOL program in the traditional form requires data to be saved to files. Because this saving to files has a few drawbacks, which is discussed later, it is customary in object-oriented programming to use databases. This chapter deals with databases, although they are not really an object-oriented feature but rather state-of-the art, describing many issues, including SQL as the most important query language for databases.

7.1 Introduction and Differences from COBOL IS Files

A database is a huge data repository

COBOL, or at least traditional COBOL (more about this later), uses individual files to store data. A program works mainly with several files open at the same time. In contrast, a database can be thought of as a huge data repository that stores all data. This section describes the basic idea behind a database and then highlights the differences from COBOL. It also explains special database functions, such as recovery or concurrency, and introduces a new generation of object-oriented databases. Finally, this chapter describes the change from ISAM COBOL...